swyx 估算 DeepSeek V4.1 Flash 推理服务约 80% 毛利
deepseek posted way more alpha on their zhihu page, if you only read twitter you are reading the laowai approved stuff
swyx 指出 DeepSeek 在知乎披露了更多细节：基于自研框架 DeepInfer 为 DeepSeek V4.1 Flash 搭建高吞吐低延迟大规模推理服务，单集群吞吐达 155 trillion tokens/day，按非高峰价格计算每卡每小时收入 38 美元。他按每 GPU 小时约 5-10 美元的成本估算，认为 DeepSeek 按其定价可做到约 80% 的服务毛利。他同时提到 DeepSeek 已开源 DeepInfer 推理框架。
查看原帖🚀 Introducing DeepInfer, our in-house inference framework, now fully open source! Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable. 🔗 https://t.co/JGAik1kgk0 1/4
正文
当前显示原文，中文全文译文尚未就绪。
deepseek posted way more alpha on their zhihu page, if you only read twitter you are reading the laowai approved stuff
"Based on this framework, we have implemented a high-throughput, low-latency large-scale inference service for DeepSeek V4.1 Flash. Currently, the service throughput of a single cluster is 155 trillion tokens/day, and the overall revenue is $38 per card per hour (based on off-peak prices)."
it costs roughtly $5-10 per gpu hour so depseek is claiming 80% serving margin at THEIR prices lololololol